Already a DIA Member? Sign in. Not a member? Join.

Sign in

Forgot User ID? or Forgot Password?

Not a Member?

Create Account and Join

The Westin San Diego Bayview Hotel

Oct 16, 2025 7:00 AM - Oct 17, 2025 7:00 PM

400 W Broadway, San Diego, CA 92101, USA

Real-World Evidence Conference

Translating Insights into Real-World Value

Hosted Session/Non-CE: Case Study hosted by Castor: On the Road to Self-Driving Trials: Faster, Cleaner RWE Data Extraction through AI Agents

Session Chair(s)

  Sponsored Sessions

Sponsored Sessions

United States

Data for observational studies still arrives as unstructured EHR and PDF records—slow, error-prone, and costly to process. Castor Catalyst is an AI agent that pulls regulatory-grade variables from any medical record, then flags exceptions for human review. In a 54-chart validation it cut extraction errors from 6.6 % to none detected, trimmed review time from 30 min to 6 min, and dropped direct cost from ˜ $49 to ˜ $10 per chart. Five live RWE studies now run Catalyst in production; this session shares what worked, what broke, and what’s next.

Featured Topics:

  • Why AI: Catalyst’s extraction beats manual work on error, speed, and cost—numbers from five live RWE studies
  • Human supervision stays: Catalyst flags every change for investigator sign-off, meeting ICH E6(R2) expectations
  • Guardrails that matter: no autonomous endpoint data, no direct patient contact, every action auditable
  • What we’ve learned so far: error rates, time-to-complete, and user-satisfaction data—where Catalyst shines, where it still needs help
  • 2026 roadmap: plug-and-play with other systems, add new “skills,” and use predefined metrics before loosening autonomy any further
  • Speaker(s)

    Lisa  Charlton

    Speaker

    Lisa Charlton

    Castor, United States

    Chief Product Officer

    Have an account?

    Be informed and stay engaged.

    Don't miss an opportunity - join our mailing list to stay up to date on DIA insights and events.